The only rule ive noted is that you can raise or reduce the levels of the audio of the items as a whole & not experience much quality loss.But the quality is compromised if you completely remove a component. I perform several conversions using small steps & split the stereo channel to mono to minimise phasing & artifacts. But the best versions are the ones which give you as much control as possible over the conversion. There are many more which perform the same task.

I think some websites use the same algorithm to convert a file to two tracks Mvsep is a free online version but it only converts to mp3 Phonicmind vocal remover is powerful but it's online subscription rates cost more than a digital wav file. You also have many online versions where you upload your file to a server & it converts them.They are usually related to karaoke. Izotope Rx7 has a very powerful feature named music rebalance but it's expensive.

There's a Vst version named regroover which which will split a drum file into hi-hats. It's available in Acid pro but you only get three converted audio tracks & you have no control over the results.

The Stem extraction feature can now be found in Djay Pro AI & also Virtual dj 2021.The dj software is real time & can isolate hi hats & snare drums.( I have not tried any of them as I use Izotope Rx7 ) I hope this tech become a common feature in every Daw.

Removing vocals spectralayers pro